Symptom‐based portopulmonary hypertension screening questionnaire in Japanese patients with chronic liver disease

Abstract Background and Aim As the exact prevalence of portopulmonary hypertension (PoPH) and the etiology of chronic liver disease (CLD) remain unknown, the present study aimed to clarify these points in Japanese patients with CLD using symptom‐based questionnaire screening. Methods Patients with CLD were asked to complete an eight‐item written questionnaire on pulmonary hypertension (PH) symptoms. If at least one item response was “yes,” the patient was offered ultrasonic echocardiography (UCG). Patients identified as having an intermediate or high risk of PH by UCG were referred to a cardiologist for further evaluation, whereby a definitive diagnosis of PoPH was made using right heart catheterization (RHC) findings. Results A total of 1111 patients with CLD completed the survey. Of the 566 symptomatic patients with at least one question answered as “yes,” approximately half agreed to undergo UCG (n = 267). Compared with asymptomatic patients, symptomatic patients were significantly older, predominantly female, and more frequently exhibited cirrhosis. Based on UCG findings, 228, 12, and 8 patients had a low, intermediate, or high risk for PH, respectively. Intermediate‐/high‐risk patients showed significantly more advanced disease progression status than low‐risk patients. The frequencies of answer to the eight questions were comparable. Ultimately, three patients were diagnosed as having PoPH (1.1% of UCG cases), one with underlying hepatitis C virus (HCV) infection and two with primary biliary cholangitis (PBC). Conclusion This symptom‐based PoPH screening study clarified the prevalence of PoPH in CLD patients according to a PH symptom questionnaire, UCG, and RHC. Patients with HCV and PBC may have a higher risk of PoPH.


Introduction
Portopulmonary hypertension (PoPH) is defined as pulmonary arterial hypertension (PAH) complicated with portal hypertension, mainly due to chronic liver disease (CLD). 1,2 PoPH occurs in 5-15% of patients with PAH 3,4 and is reportedly found in 2-6% of patients with portal hypertension and 1-2% patients with liver cirrhosis. 3,5 In contrast to Europe and the United States, the real-world status of PoPH in Japan is largely unknown. 3,4 Atsukawa et al. retrospectively analyzed 186 patients simultaneously undergoing hepatic vein and pulmonary artery catheterization to investigate the prevalence of PoPH and identified two patients (1%) with the disorder. However, the exact prevalence of PoPH remains unclear in Japanese patients with CLD, with the low prevalence suggesting underestimation and an incomplete clinical picture of PoPH.
The pathogenesis of PoPH is uncertain. In addition to genetic predisposition, 6 thromboembolism in the portal venous system, 7 inflammation, 8 hyperdynamic pulmonary circulation, 9 and imbalances in vasoconstrictive and vasodilatory mediators due to reduced liver metabolism 10 have all been associated with PoPH development. Indeed, higher rates of PoPH were reported in patients with end-stage liver disease undergoing liver transplantation, although the prevalence of PoPH appeared uninfluenced by the severity of liver disease. 4,10 Whereas no gender differences have been associated with PoPH prevalence, 11 women with autoimmune hepatitis (AIH) may have a higher risk of PoPH onset. 12 Accumulating evidence also suggests that the existence of CLD, including AIH, hepatitis C virus (HCV) infection, and alcoholic liver disease (ALD), are linked to PoPH development. 11 However, it is unknown precisely which CLDs are most involved in PoPH onset in the Japanese.
The clinical manifestations of PoPH are identical to those of pulmonary hypertension (PH), that is, dyspnea on exertion, atypical chest pain, elevated jugular venous pressure, leg edema, and others, 1,13 all of which are considered nonspecific symptoms of PoPH. Therefore, the threshold for suspecting PoPH is presumably low since patients with CLD can exhibit dyspnea for a variety of reasons. A simple but accurate medical history arrived at through an interview is needed for identifying at-risk candidates for further examination with ultrasonic echocardiography (UCG) before the invasive final diagnostic procedure of right heart catheterization (RHC). Moreover, as complicating PoPH in CLD patients is a poor prognostic factor in spite of targeted therapy intervention, 11 early detection and treatment is needed to improve prognosis in patients with PoPH.
Guidelines for pulmonary hypertension recommend echocardiography for screening of symptomatic patients, but it is not clear which patients should be screened for UCG because the clinical presentation of PoPH is not clear. The present study aimed to address to characterize patients with PH suffering from CLD and to identify UCG recommendations in patients with CLD through initial screening with a novel questionnaire based on PH symptoms.

Methods
Study design. This protocol of this prospective, single-center, observational study was reported elsewhere and registered as UMIN 000042287 on 29 October 2020. 14 The inclusion criteria for analysis were as follows: (i) over 20 years of age, (ii) having CLD, and (iii) displaying at least one symptom suggestive of PH based on an eight-item questionnaire (described below). Patients who were unable to be followed at our hospital were excluded. This study was reviewed and approved by the Institutional Review Board of Shinshu University School of Medicine (no. 4891) on 29 September 2020, and was conducted according to the principles of the Declaration of Helsinki.
Patients. A total of 1111 patients with CLD who visited Shinshu University Hospital (Matsumoto, Japan) between 16 November 2020 and 16 October 2021, were prospectively registered in this study and included for observational analysis.
Study flow. The study flowchart is presented in Figure 1. While waiting for medical examination in the outpatient clinic of Shinshu University Hospital, the participants completed an eightitem questionnaire written in Japanese, translated as follows: Q1: Do you feel you cannot work at an intensity comparable to others of the same age and gender? Q2: Do you feel you cannot move as fast as others of the same age and gender? Q3: Do you feel you cannot move at the same pace as others of the same age and gender? Q4: Do you need rest when climbing stairs or carrying heavy loads? Q5: Do you ever experience shortness of breath? Q6: Do you ever experience faintness? Q7: Do you ever experience tiredness and/or persistent malaise? Q8: Do you ever experience facial edema and/or pretibial edema?
These questions are based on the symptoms of PH. 15,16 If at least one question was answered as "yes," the attending physician advised the patient to undergo UCG as a first-line screening for PoPH and requested written informed consent for participation in the study. Consenting patients received UCG evaluation, the cost of which was covered by the national health insurance system. If the participant exhibited UCG-positive findings indicating an intermediate or high probability of PoPH based on published guidelines 15,16 (presented in Table S1A,B, Supporting information), he or she was referred to the Department of Cardiovascular Medicine for further evaluation, whereby a definitive diagnosis was made using RHC findings. 15,16 Etiology of CLD. The etiologies of CLD considered in this study were the following: (i) hepatitis B virus (HBV) infection defined as HBe antigen-positive asymptomatic carriers, chronic hepatitis, HBe antigen-negative inactive carriers, and HBs antigennegative clinical remission regardless of cirrhosis status according to the guidelines of the Japan Society of Hepatology (JSH) 17 ; (ii) HCV infection including patients with persistent HCV infection and with post-HCV eradication in a sustained virological response (SVR) based on the guidelines of the JSH 18 ; (iii) ALD including patients with a drinking habit (ethanol >60 g/day for men and >40 g/day for women), and persistent elevation of aminotransferase and γ-glutamyltranspeptidase according to the guidelines of the Japanese Society of Gastroenterology (JSGE) 19 ; (iv) nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH) according to the guidelines of the JSH and the JSGE 20 ; and (v) autoimmune liver disease, including AIH, primary biliary cholangitis (PBC), and primary sclerosing cholangitis (PSC) based on the JSH guidelines. [21][22][23] Diagnosis of clinical stage. Liver cirrhosis was diagnosed by histological examination and/or characteristic clinical signs of advanced liver disease including noninvasive test such as FIB-4 index. Hepatocellular carcinoma (HCC) was determined by histological examination, blood test, and/or imaging studies. The presence of an esophagogastric varices was determined by computed tomography (CT), magnetic resonance imaging, and/or esophagogastric endoscopy. The presence of portal hypertension was defined by the measurement of hepatic venous pressure gradient (HVPG), or by the presence of cirrhosis, the complication of esophagogastric varices, or the presence of a portal-systemic shunt for cases in which it was difficult to measure HVPG.
Statistical analysis. Statistical analysis and data visualization were carried out using StatFlex ver. 7.0.11 software (Artech Co., Ltd., Osaka, Japan) and Python (version 3.9), scipy (1.9.3), and Scikit-learn (version 1.2.0). Data are presented as the median AE interquartile range (IQR) for continuous variables. Continuous variables were statistically evaluated by means of the Mann-Whitney U test, while categorical variables were analyzed using the chi-square test. All statistical tests were two-sided and evaluated at the 0.05 level of significance. Figure 1 Study flowchart. † Pulmonary hypertension (PH) low-risk group included one symptomatic patient who was previously diagnosed as having portopulmonary hypertension (PoPH) + connective tissue-related pulmonary artery hypertension (CTD-PAH), had already been treated by tadalafil and macitentan, and had improved ultrasonic echocardiography parameters (Case No. 3 in Table 5). ‡ Of the two cases of PAH, one was from an unknown cause and the other was due to lung disease.
Questionnaire screening results. Among the eight questions used in PH screening, the most frequent symptom reported was "Q3: Do you feel you cannot move at the same pace as others of the same age and gender?" in 325 cases (29.3%) ( Figure S1). A total of 588 patients (52.9%) answered "yes" to at least one symptom of suspected PH and were classified as symptomatic patients.
Comparisons of clinical characteristics between symptomatic and asymptomatic patients. The clinical characteristics of the symptomatic patients (n = 588) are summarized in Table 2. Compared with asymptomatic patients, they were significantly older (P < 0.0001), female (P = 0.0031), and had lower height (P < 0.0001) and weight (P = 0.00314). The frequencies of HCV (P = 0.0066), liver cirrhosis (P < 0.0001), history of HCC (P = 0.0085), and history of esophagogastric varices (P = 0.0085) were all significantly higher in the symptomatic group. In addition, symptomatic patients were significantly more likely to have a history of Patients with liver cirrhosis were significantly more likely to show symptoms of PH than non-cirrhotic patients (P < 0.0001) ( Table 3). We observed that 68.9% of cirrhosis patients had at least one PH symptom.
The incidence of almost all symptoms apart from "Q5: Do you ever experience shortness of breath?" was significantly higher in liver cirrhosis patients (P < 0.01 each). The most common symptom was "Q3: Do you feel you cannot move at the same pace as others of the same age and gender?," which was positive in 42.8% of patients.
Clinical characteristics of suspected pulmonary hypertension patients. UCG was performed on 267 consenting patients. Table 4 shows comparisons of clinical characteristics between symptomatic patients with or without UCG. Twelve (4.5%) and 10 (3.7%) patients had an intermediate risk or high risk of PH, respectively, based on UCG results (Fig. 1).
Patients with intermediate/high risk (n = 22; 8.2% of symptomatic patients) had significantly higher rates of a history of HCC (P = 0.0085) and of varices (P = 0.0085) versus low-risk patients (n = 245) ( Table 5). White blood cell count, hemoglobin, and platelet count were significantly lower, while lactate dehydrogenase (LDH) and blood urea nitrogen (BUN) were significantly higher, in intermediate-/high-risk patients (  (Fig. 1, Table S2). Finally, as shown in Figure 1 and Table 6, case 1 with high-risk UCG findings was newly diagnosed in this study after further evaluation by RHC, whose etiology was HCV-related liver cirrhosis. Cases 2 and 3 were previously diagnosed as having PoPH and had  already been treated by tadalafil and/or macitentan. Both had PBC at the non-cirrhotic stage but exhibited portosystemic shunts and splenomegaly. Since both patients had systemic sclerosis, they were judged as having PoPH with concomitant connective-tissue-disease-associated PAH (CTD-PAH). Specifically, one of the two PBC patients showed improvements in UCG parameters by PAH treatment, although several symptoms remained. Therefore, three patients were ultimately diagnosed as having PoPH (1.1% of UCG cases) as summarized in Figure 1 and Table 6. Our study also revealed two patients complicated with PAH and 10 patients complicated with PH due to other causes (Fig. 1).

Discussion
This PH symptom-based screening study confirmed that roughly one-half of CLD patients (n = 1111) had at least one symptom related to PH (n = 588). Symptomatic patients tended to be older, be female, and have a more advanced disease status, including liver cirrhosis, versus asymptomatic patients. Twenty-two of 267 patients (8.2%) were categorized as with intermediate/high risk of PH by UCG. Ultimately, our study identified one HCV-associated case and two PBC-associated cases already being treated for PAH as having PoPH (1.1% of UCG cases). These findings shed light on the prevalence and etiology of PoPH in Japan.
The final diagnosis of PH including PoPH is based on RHC. Because of its invasive nature, however, it is difficult to perform RHC for all suspected cases of PH in the clinical setting. UCG is the recommended first step for PH screening by the European Society of Cardiology guidelines and the Japanese Circulation Society guidelines. 15 Because performing UCG for all patients with CLD is infeasible from a financial standpoint, a simple preliminary screening system is needed. A delayed diagnosis is reportedly associated with poor PH prognosis. 24 However, a previous study revealed that 21% of patients had symptoms for over 2 years before PH diagnosis, 25 confirming the need for appropriate noninvasive screening methods such as the one in the present investigation. Indeed, roughly half of patients with CLD had symptoms indicative of PH and were indicated for UCG analysis. Patients at the cirrhosis stage in CLD frequently have nonspecific clinical manifestations, including fatigue, anorexia, and weight loss. Therefore, positive symptoms for PH were expected; in fact, 68.9% of cirrhosis-stage CLD patients had at least one PH manifestation. Most patients with compensated cirrhosis or chronic hepatitis are considered to be asymptomatic for PH. However, half of the cohort displayed one or more PH symptoms, suggesting that such patients may indeed present with PH-like findings. Taken together, the eight questions included in this study on CLD patients may be a clue to the screening method for PH candidate detection. Further investigation is needed to clarify whether asymptomatic patients are complicated with PH.
Regarding biomarkers for PH detection, abnormalities in blood parameters such as brain natriuretic hormone (BNP) and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) are not useful for the early detection of PH since those values increase with disease severity. 26,27 Very recently, it was reported that CT-based measurements of main pulmonary artery diameter (mPA-D) and mPA-D/ascending aorta diameter (aAo-D) could identify patients with possible PoPH in clinical practice focused on portal hypertension. 28 Moreover, another study found the ALBI score to be the most impacted factor of severe PoPH and potentially useful for the estimation of pulmonary vascular resistance. 29 Because of the limited number of PoPH patients detected in our cohort, the efficacy of all proposed methods should be validated in future studies of larger cohorts.
In terms of PoPH etiology, three cases were identified at the final diagnosis in our study with backgrounds of HCV or PBC. Unlike cohorts from Western countries, in which the proportion of ALD is high, 30 viral hepatitis-related CLD is more prevalent in Japan. Kawaguchi et al. recently reported HCV with or without an SVR as a major etiology of chronic liver disease in patients with PoPH. 29 They suggested that all HCV patients should be screened for PoPH, regardless of the SVR status.
Several studies on PoPH in Asia and Japan investigating disease prevalence and incidence have shown that AIH and PBC patients were more likely to be complicated with PoPH. 5,31 Indeed, we encountered two patients with PBC previously diagnosed as having CTD-PAH who had already been treated. AIH    and PBC have a high complication rate of CTD. 32,33 CTD can also be a cause of PAH as the second most prevalent type of PAH, and systemic sclerosis is the most prevalent type of CTD. 24 In PBC patients, systemic sclerosis is the third most common concomitant autoimmune disease (1.4-12.3%). 32 As in our cases, autoimmune liver disease can coexist with CTD-PAH to harbor a high risk of PoPH development. PoPH screening by questionnaires and UCG may therefore be important not only in patients with cirrhosis but also in symptomatic patients with chronic hepatitis, particularly those with autoimmune liver disease such as AIH and PBC. The prevalence of PoPH in UCG examinations was 1.1%, which was similar to a previously reported estimate of 1-2% among liver cirrhosis patients. 3 However, given that there were 49 cases of liver cirrhosis out of 267 cases, there might be a relatively high proportion of PoPH among liver cirrhosis patients. This discrepancy suggested that the prevalence of PoPH was not influenced by liver disease severity. 4,10 In addition, a higher proportion of deaths due to PoPH has been reported in patient groups with good liver function. 34 Clinicians should therefore pay attention to both cirrhosis and non-cirrhosis patients for earlier PoPH detection. 34 There are several limitations to this investigation. First, it was a single-center cohort study. Second, some patients, including elderly ones, refused further evaluation due to the invasiveness of RHC and did not reach a final diagnosis, thus supporting the necessity of determining which patients will benefit from RHC. Third, we could not completely rule out the possibility of asymptomatic PoPH patients having PoPH. However, the presence of symptoms is the main criterion for disease identification based on the diagnostic guidelines for PH, with 86% of PAH cases reportedly exhibiting symptoms at the time of diagnosis. 25 Therefore, we believe that our tool may be particularly useful to focus on such symptomatic cases towards UCG. In conclusion, this symptom-based PoPH screening study detected three PoPH patients at a prevalence of 1.1% among symptomatic patients screened by a PH symptom-based questionnaire, UCG, and RHC. Patients with a background of HCV or PBC may have a risk of PoPH and should be closely monitored for PH symptoms.